How to use molfile_to_params_polymer.py to create CRO.params?

[lanselibai]

Many thanks to Rocco Moretti's answer that CRO.params already exists and can be used by
-extra_res_fa ${ROSETTA}/database/chemical/residue_type_sets/fa_standard/residue_types/extra/CRO.params
#383

I would like to learn how to create the CRO.params myself as I have other chromophores to parameterise.

It looks like molfile_to_params.py will output TYPE LIGAND and without having LOWER_CONNECT N, UPPER_CONNECT C, ICOOR_INTERNAL UPPER, ICOOR_INTERNAL LOWER lines in the .params file.

I tried molfile_to_params_polymer.py with this command:
python ~/rosetta.binary.ubuntu.release-371/main/source/scripts/python/public/molfile_to_params_polymer.py -i CRO.mol2 --polymer -n CRO -p CRO

It correctly outputs lines of:

TYPE POLYMER
LOWER_CONNECT N
UPPER_CONNECT C

But got an error of the below and without ICOOR_INTERNAL lines in the CRO.params. Can I ask how to resolve this?

$ python ~/rosetta.binary.ubuntu.release-371/main/source/scripts/python/public/molfile_to_params_polymer.py -i CRO.mol2 --polymer -n CRO -p CRO
No valid centroid coords are provided, calculate centroid manually
Centering ligands at (  41.410,  -23.601,    2.368)
WARNING:  atom C2 has valence > 4
the molecule has 37 atoms
Total naive charge -2.525, desired charge 0.000, offsetting all atoms by 0.068
WARNING: fragment 1 has 37 total atoms including H; protein residues have 7 - 24 (DNA: 33)
WARNING: fragment 1 has 9 rotatable bonds; protein residues have 0 - 4
Average 37.0 atoms (22.0 non-H atoms) per fragment
(Proteins average 15.5 atoms (7.8 non-H atoms) per residue)
WARNING:  no root atom specified, using auto-selected NBR atom instead.
Preforming polymer modifications
REVERSED!!!
Traceback (most recent call last):
  File "/home/lslb/rosetta.binary.ubuntu.release-371/main/source/scripts/python/public/molfile_to_params_polymer.py", line 300, in <module>
    sys.exit(main(sys.argv[1:]))
             ^^^^^^^^^^^^^^^^^^
  File "/home/lslb/rosetta.binary.ubuntu.release-371/main/source/scripts/python/public/molfile_to_params_polymer.py", line 236, in main
    write_poly_param_file(param_file, m, args.name, 1, args.peptoid, args.use_parent_rotamers)
  File "/home/lslb/rosetta.binary.ubuntu.release-371/main/source/scripts/python/public/molfile_to_params_polymer/IO_functions.py", line 515, in write_poly_param_file
    nbb.input_stub1,  nbb.input_stub2,  nbb.input_stub3  = nbb, cabb, cbb
    ^^^^^^^^^^^^^^^
AttributeError: 'int' object has no attribute 'input_stub1'

CRO.mol2.txt

CRO_wrong.params.txt

CRO.params.txt

@vmullig
@dougrenfrew
@tlinsky

[roccomoretti]

molfile_to_params_polymer.py isn't really set up well to handle non-alpha amino acids. (And molfile_to_params.py will treat it as a free ligand, not as a polymeric residue.) That's what PR #419 is intending to fix.

And indeed, CRO is the example chosen for the integration test - you can take a look at the command file & associated input file in the PR to see how it's set up. But that will only work if you check out the branch, or if you wait until the PR gets merged.